Sto provando a rilevare un oggetto dal seguente codice che coinvolge il rilevatore di onde, non voglio disegnare corrispondenze, voglio disegnare un rettangolo attorno all'oggetto rilevato, ma in qualche modo non riesco a ottenere l'omografia corretta, per favore qualcuno può indicare dove sto andando male.Disegno rettangolo attorno all'oggetto rilevato usando SURF
#include <stdio.h>
#include <iostream>
#include "opencv2/core/core.hpp"
#include "opencv2/features2d/features2d.hpp"
#include "opencv2/highgui/highgui.hpp"
#include "opencv2/imgproc/imgproc.hpp"
#include "opencv2/calib3d/calib3d.hpp"
using namespace cv;
int main()
{
Mat object = imread("sample.jpeg", CV_LOAD_IMAGE_GRAYSCALE);
if(!object.data)
{
std::cout<< "Error reading object " << std::endl;
return -1;
}
//Detect the keypoints using SURF Detector
int minHessian = 500;
SurfFeatureDetector detector(minHessian);
std::vector<KeyPoint> kp_object;
detector.detect(object, kp_object);
//Calculate descriptors (feature vectors)
SurfDescriptorExtractor extractor;
Mat des_object;
extractor.compute(object, kp_object, des_object);
FlannBasedMatcher matcher;
VideoCapture cap(0);
namedWindow("Good Matches",0);
cvResizeWindow("Good Matches",800,800);
std::vector<Point2f> obj_corners(4);
//Get the corners from the object
obj_corners[0] = (cvPoint(0,0));
obj_corners[1] = (cvPoint(object.cols,0));
obj_corners[2] = (cvPoint(object.cols,object.rows));
obj_corners[3] = (cvPoint(0, object.rows));
char key = 'a';
int framecount = 0;
while (key != 27)
{
Mat frame;
cap >> frame;
if (framecount < 5)
{
framecount++;
continue;
}
Mat des_image, img_matches;
std::vector<KeyPoint> kp_image;
std::vector<vector<DMatch > > matches;
std::vector<DMatch > good_matches;
std::vector<Point2f> obj;
std::vector<Point2f> scene;
std::vector<Point2f> scene_corners(4);
Mat H;
Mat image;
cvtColor(frame, image, CV_RGB2GRAY);
detector.detect(image, kp_image);
extractor.compute(image, kp_image, des_image);
matcher.knnMatch(des_object, des_image, matches, 2);
for(int i = 0; i < min(des_image.rows-1,(int) matches.size()); i++) //THIS LOOP IS SENSITIVE TO SEGFAULTS
{
if((matches[i][0].distance < 0.6*(matches[i][1].distance)) && ((int) matches[i].size()<=2 && (int) matches[i].size()>0))
{
good_matches.push_back(matches[i][0]);
}
}
//Draw only "good" matches
// drawMatches(object, kp_object, image, kp_image, good_matches, img_matches, Scalar::all(-1), Scalar::all(-1), vector<char>(),DrawMatchesFlags::NOT_DRAW_SINGLE_POINTS);
if (good_matches.size() >= 4)
{
for(int i = 0; i < good_matches.size(); i++)
{
//Get the keypoints from the good matches
obj.push_back(kp_object[ good_matches[i].queryIdx ].pt);
scene.push_back(kp_image[ good_matches[i].trainIdx ].pt);
}
H = findHomography(obj, scene, CV_RANSAC);
perspectiveTransform(obj_corners, scene_corners, H);
//Draw lines between the corners (the mapped object in the scene image)
line(image, scene_corners[0] + Point2f(object.cols, 0), scene_corners[1] + Point2f(object.cols, 0), Scalar(0, 255, 0), 4);
line(image, scene_corners[1] + Point2f(object.cols, 0), scene_corners[2] + Point2f(object.cols, 0), Scalar(0, 255, 0), 4);
line(image, scene_corners[2] + Point2f(object.cols, 0), scene_corners[3] + Point2f(object.cols, 0), Scalar(0, 255, 0), 4);
line(image, scene_corners[3] + Point2f(object.cols, 0), scene_corners[0] + Point2f(object.cols, 0), Scalar(0, 255, 0), 4);
}
//Show detected matches
imshow("Good Matches", image);
key = waitKey(1);
}
return 0;
}
in qualche modo in grado in che modo? non ottieni abbastanza buoni risultati, l'omografia calcolata è semplicemente sbagliata? forse hai qualche partita fuori dagli oggetti? – penelope
@penelope le corrispondenze sono abbastanza buone, ma il rettangolo che dovrebbe essere intorno all'oggetto non è intorno ad esso, quindi penso di calcolare le coordinate sbagliate. – sum2000